GFS Performance / Government Shutdown

Hi there.

I’ve just been reading an article regarding the GFS and the government shutdown.

Washington Post reporting that GFS performance is being affected due to lack of employees maintaining how it works. I don’t know whether it’s fake news or not, but it was mentioned on the weather channel, so who knows?

https://www.washingtonpost.com/weather/2019/01/07/national-weather-service-is-open-your-forecast-is-worse-because-shutdown

It’ll be interesting to monitor WXSIM’s output. :smiley:

Regards.

If a person (really) gives a rats tail about accuracy, such a person be very cautious about the data said person would ingest.

99% of your country uses GFS data for official forecasts. :lol:

I’m not sure that’s still true :? Even my local TV station’s 2 man wx operation looks at multiple models, and if there’s a big divergence between GFS and e.g. ECMWF then they show it both ways.

They don’t all rely a single model, they’ll use a blend of data. But you can bet that the GFS data is almost certainly always used within that blend for your country - it’s your premier model. The same way our Met Office uses a blend of ECM, DWD and MOGREPS data.

Agreed, a blend it is, but if the GFS is an outlier its data is likely to be discounted in the end result.

Thanks Niko, the above is what I was attempting to articulate. :smiley:

Well, it seems to confirm my original post.

The automation of the GFS data has changed due to the shutdown. This means that the initialisation has issues, and the human input for removing/adding in bias to the model is not being done. This also means that work has stopped on the F3V model too - it hasn’t even initialised the 12Z. This will also impact the FV3 taking over as the GFS operational, a substantial delay is likely.

Verification stats for both the GFS and FV3 are extremely poor right now.